NapMem: A Framework for Active Memory Navigation in Conversational AI Agents
By
[Submitted on 7 Jul 2026]
Summary
This paper introduces NapMem, a framework that transforms long-term user memory in conversational AI from a passive retrieval system into a structured action space. Instead of treating memory as pre-selected context fed to a model, NapMem organizes user history into a multi-granularity memory pyramid (raw conversations, typed records, topic tracks, user profiles) connected through provenance relations. The agent is trained via reinforcement learning to actively navigate and select memory at appropriate granularities based on queries and intermediate evidence. Experiments on benchmarks (PersonaMem-v2, LongMemEval, LoCoMo) show competitive performance on memory-intensive tasks while preserving general reasoning abilities.
Source
Twitter / XNapMem: A Framework for Active Memory Navigation in Conversational AI Agentsarxiv.orgKey quotes
· 5 pulledWe introduce NapMem, a framework for learning to use long-term user memory as a structured action space rather than passively retrieved context.
NapMem organizes user history into a linked multi-granularity memory pyramid, where raw conversations, typed memory records, topic tracks, and user profiles are connected through provenance relations.
The agent is trained to select memory according to the query and intermediate evidence, allowing it to inspect different memory granularities before answering.
Experiments on PersonaMem-v2, LongMemEval, and LoCoMo show that a NapMem agent trained with memory-tool reinforcement learning is competitive across diverse memory-intensive tasks.
Our results suggest that long-term user memory benefits from coupling structured storage with a learned policy for using memory at the appropriate granularity.
You might also wanna read
Robot Memory System Enables AI Robots to Learn from Past Experiences
Robot Memory (robotmem) is a persistent memory system for AI robots that enables robots to learn from past experiences. The system stores ep
github.com·4mo ago
Memori launches agent-native persistent memory infrastructure using structured knowledge graphs from agent trace data
Memori is a new agent-native memory infrastructure that enables AI agents to create structured, long-term persistent memory directly from ag
Product Hunt·2mo ago
Building Production AI Agents: A Multi-Tiered Memory Architecture with Zep, Mem0, and ContextNest
This article discusses the need for a multi-tiered persistent memory architecture when building production-grade AI agents. It argues that r
promptowl.ai·5d ago
MemPO: Self-Memory Policy Optimization Algorithm Improves Long-Horizon Agent Performance While Reducing Token Usage
This paper introduces MemPO (Self-Memory Policy Optimization), a novel algorithm that enables long-horizon AI agents to autonomously summari
arxiv.org·21d ago
JAMEL: A Framework for Joint Memory and Exploration Learning in Language Model Agents
This paper introduces JAMEL (Joint Agent Memory and Exploration Learning), a framework that trains language model agents to explore open-end
arxiv.org·1mo ago
δ-mem: A Compact Online Memory Mechanism for Efficient Long-Context LLM Processing
The article presents δ-mem, a lightweight memory mechanism for large language models that augments frozen full-attention backbones with a co
arxiv.org·1mo ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.